# The latest in managing and auditing GitHub Copilot agents Source: https://www.youtube.com/watch?v=boviC841YWs Recap page: https://rapidrecap.app/video/boviC841YWs Generated: 2025-12-03T15:35:00.574+00:00 --- ## Quick Overview GitHub Copilot Enterprise now offers expanded management and auditing capabilities, allowing administrators to create granular custom enterprise roles specifically for AI controls, define agent configuration rules using code rulesets that apply enterprise-wide, and track all AI agent activities via an audit log filtered by actor:Copilot. **Key Points:** - GitHub introduced new tools for controlling Copilot data and access, including expanding agent configuration and access control capabilities for enterprises. - Administrators can create custom enterprise roles, such as 'AI Admins' (00:54), and assign specific permissions like 'Manage enterprise AI controls' (01:01). - Custom AI agents, defined via Markdown files in a repository like 'github-private/agents' (01:39), can be made available across the entire enterprise by selecting the source organization (01:56). - Enterprise-level push rulesets can be created under Policies (02:05) to restrict changes to agent configuration files, such as limiting pushes that modify files in the 'agents/**/*.md' path (02:19). - The Audit log automatically filters events where the actor is 'Copilot' (02:36), showing detailed JSON event data for actions like pull request creation (02:50). - The Model Context Protocol (MCP) settings allow administrators to enable MCP servers everywhere or restrict access to only those listed in the enterprise's registry (03:36, 03:47). ![Screenshot at 00:16: The administrator navigates to the 'Copilot usage' insight section, displaying key metrics like IDE active users \(508\) and Agent adoption rate \(82%\) for the current calendar month.](https://ss.rapidrecap.app/screens/boviC841YWs/00-00-16.png) **Context:** This video details recent enhancements to GitHub Copilot's enterprise management features, focusing on granular control over AI agents, Copilot usage, and security auditing. The demonstration moves through the Copilot Insights dashboard, enterprise role management, AI Controls for agents, Policy settings for rulesets, and finally, the Audit Log, illustrating how administrators gain control over Copilot's integration and operation within a large organization. ## Detailed Analysis The presentation showcases significant updates to GitHub Copilot Enterprise management and control features. It begins by reviewing the Copilot Usage Insights dashboard (00:00), which displays metrics like 508 active users and 82% agent adoption. The focus quickly shifts to governance, demonstrating how administrators can create custom enterprise roles, such as 'AI Admins' (00:49), and assign them specific permissions, like managing enterprise AI controls (00:59). Next, the video covers the creation of custom AI agents using configuration files stored in a repository (01:36), which can then be enabled enterprise-wide via the AI Controls Agent settings (01:56). A crucial governance feature shown is the ability to create custom Code Rulesets under Policies (02:05); specifically, a push ruleset is configured to restrict file modifications within the agent configuration directory, such as blocking changes to 'agents/**/*.md' files (02:19). Finally, the demonstration moves to the Audit Log (02:32), where filtering by 'actor:Copilot' reveals detailed JSON logs of every AI action taken, including pull request creations (02:50), confirming who performed what action and where. The video concludes by reviewing the MCP settings, allowing admins to control whether developers can use external MCP servers or restrict them to only those listed in the enterprise's registry (03:36). ### Copilot Usage Insights - 508 IDE active users this month - 82% agent adoption rate - Claude Sonnet 4 is the most used chat model ### Enterprise Role Management - Custom roles like 'AI Admins' can be created (00:49) and assigned permissions like 'Manage enterprise AI controls' (01:01) ### Custom Agent Configuration - Agents are defined in Markdown files within a source repository (01:39) and made available enterprise-wide by selecting the source organization (01:56) ### Policy Enforcement via Rulesets - New push rulesets can restrict pushes that modify specific agent configuration files, demonstrated by restricting changes to 'agents/**/*.md' paths (02:17) ### Activity Auditing - The Audit log automatically filters events where actor='Copilot' (02:36), providing granular details on actions like pull request creation (02:50) including the user, agent, and repository involved ### Model Context Protocol (MCP) Controls - Admins set MCP servers to 'Enabled everywhere' or restrict access to 'Registry only' based on the defined MCP registry URL (03:36) ![Screenshot at 00:00: The main Copilot ID usage dashboard showing key metrics like active users, agent adoption, and the most used chat model.](https://ss.rapidrecap.app/screens/boviC841YWs/00-00-00.png) ![Screenshot at 00:51: The Enterprise Role Management screen where an administrator creates a new custom role named 'AI Admins' \(00:54\).](https://ss.rapidrecap.app/screens/boviC841YWs/00-00-51.png) ![Screenshot at 01:39: Viewing the content of a custom agent definition file, 'arm-migration-agent.md', which details its purpose and execution command.](https://ss.rapidrecap.app/screens/boviC841YWs/00-01-39.png) ![Screenshot at 02:10: Creating a new Push Ruleset under Policies, specifically naming it 'Limit custom agent access' and setting it to be disabled initially \(02:11\).](https://ss.rapidrecap.app/screens/boviC841YWs/00-02-10.png) ![Screenshot at 02:35: The Audit Log screen filtered by 'actor:Copilot', displaying recent AI actions matching the search query.](https://ss.rapidrecap.app/screens/boviC841YWs/00-02-35.png)